Evaluation of compaction equations and prediction using adaptive neuro-fuzzy inference system on compressibility behavior of AA 6061 100 − x –x wt.% TiO 2 nanocomposites prepared by mechanical alloying

Abstract

Nanocrystallite/nanocomposite powders of AA 6061 100 − x –x wt.% TiO 2 (x = 0, 2, 4, 6, 8, 10 and 12) prepared by mechanical alloying and compacted at room temperature have been used for the present investigation. Compaction behavior of post-compacts as a function of compaction pressure and the nano titania content in the nanocrystallite matrix powder was investigated using several powder compaction equations (empirical form) including both linear and non-linear type. The non-linear equation proposed by Van Der Zwan and Siskens was the best fitting curve comparing other equations developed by Balshin, Heckel, Ge, Panelli and Ambrosio Filho, Kawakita, and Shapiro. The Van Der Zwan and Siskens compacting equation gives the regression coefficient very close to unity. Also, this paper focuses on the development of expert system based on an adaptive neuro-fuzzy inference system (ANFIS) on compaction behavior of the developed nanocomposite powder. This ANFIS model was accurately established to obtain the relationship between percentage of nano titania content in the nanocrystalline matrix and compaction pressure to get the required relative density. The predicted relative density obtained from ANFIS was compared with experimental data and also evaluated with the predicted relative density derived by multiple regression analysis (MRA). The comparisons indicated that the developed ANFIS achieved excellent accuracy and it was as high as 99.50%.